GOD EXAMINED10 Research Domains — 60+ Studies — Every Number
The Mathematical Case for Christianity
Every probability calculation, formal proof, statistical analysis, and empirical finding compiled into one document. The numbers speak.
The Master Evidence Table
| Evidence | Key Number | Source |
|---|---|---|
| 8 messianic prophecies by chance | 1 in 1017 | Peter Stoner, ASA reviewed |
| 48 messianic prophecies by chance | 1 in 10157 | Peter Stoner |
| Initial entropy of universe | 1 in 1010123 | Roger Penrose (Cambridge/Oxford) |
| Cosmological constant tolerance | 1 in 10120 | Multiple physicists |
| Life-supporting planet anywhere | 1 in 101032 | Hugh Ross (922 parameters) |
| Functional protein by chance | 1 in 1077 | Douglas Axe, J. Mol. Biol. 2004 |
| Functional proteome by chance | 1 in 10722,000 | Luke Barnes (astrophysicist) |
| Resurrection probability (Bayesian) | 97% | Swinburne (Oxford) |
| Resurrection Bayes factor | 1039 to 1 FOR | McGrews (Blackwell Companion) |
| Godel's ontological proof | Computer verified 2013 | Benzmuller & Paleo, Isabelle/HOL |
| P(God|all evidence) | > 50% | Swinburne (Oxford cumulative) |
| P(God) independent calculation | 67% | Unwin (physicist) |
| NT Greek manuscripts | 5,800+ | Next closest: Homer at 1,757 |
| Total NT manuscripts | 25,000+ | All languages combined |
| Isaiah accuracy over 1,000 years | 95%+ word-for-word | Dead Sea Scrolls vs Masoretic |
| Biblical persons confirmed | 53 | Mykytiuk, Purdue / BAR |
| Scholars: Jesus existed | ~99% | Habermas, 3,400 publications |
| Scholars: disciples had experiences | ~95% | Habermas survey |
| Suicide reduction (weekly attendance) | 5x lower | Harvard, N=89,708 |
| All-cause mortality reduction | 33% | JAMA Internal Medicine, N=74,534 |
| Life expectancy added | +7 to +14 years | Multiple studies |
| AA vs CBT abstinence | 42% vs 35% | Cochrane Review (gold standard) |
| Christian Nobel laureates | 65%+ of all prizes | Baruch Shalev, UN statistics |
| Growth rate for 300 years | 40% per decade | Rodney Stark (Princeton) |
| Daniel's 70 Weeks calculation | 173,880 days → Palm Sunday 32 AD | Anderson / British Royal Observatory |
Prophecy Probability: Peter Stoner
Stoner (Chairman, Mathematics & Astronomy, Pasadena City College / Westmont College) calculated the probability of one person fulfilling specific messianic prophecies by chance. Reviewed by the American Scientific Affiliation.
| # | Prophecy | OT Ref | Probability |
|---|---|---|---|
| 1 | Born in Bethlehem | Micah 5:2 | 1 in 280,000 |
| 2 | Forerunner prepares His way | Malachi 3:1 | 1 in 1,000 |
| 3 | Enters Jerusalem on a donkey | Zech 9:9 | 1 in 100 |
| 4 | Betrayed by a friend | Zech 13:6 | 1 in 1,000 |
| 5 | Betrayed for 30 pieces of silver | Zech 11:12 | 1 in 1,000 |
| 6 | Money buys potter's field | Zech 11:13 | 1 in 100,000 |
| 7 | Silent before accusers | Isaiah 53:7 | 1 in 100 |
| 8 | Crucified (hands/feet pierced) | Psalm 22:16 | 1 in 10,000 |
Combined: 1 in 1017 — Cover Texas 2 feet deep in silver dollars. Mark one. Blindfold, walk anywhere, pick one coin. Those are the odds.
48 prophecies: 1 in 10157 — Exceeds Borel's cosmic impossibility threshold (1050) by 107 orders of magnitude.
48 prophecies: 1 in 10157 — Exceeds Borel's cosmic impossibility threshold (1050) by 107 orders of magnitude.
Daniel's 70 Weeks: Mathematical Timeline
Daniel 9:24-27: 69 "weeks of years" = 483 years from decree to rebuild Jerusalem until "Messiah the Prince."
Anderson's calculation: 483 × 360 days = 173,880 days
Start: Nisan 1, 445 BC (Artaxerxes' decree) → End: April 6, 32 AD — Palm Sunday
Certified by the British Royal Observatory.
Anderson's calculation: 483 × 360 days = 173,880 days
Start: Nisan 1, 445 BC (Artaxerxes' decree) → End: April 6, 32 AD — Palm Sunday
Certified by the British Royal Observatory.
Fine-Tuning: The Constants
| Constant | Tolerance | What Happens If Wrong |
|---|---|---|
| Cosmological constant | 1 in 10120 | Universe flies apart or collapses |
| Gravity vs cosmological constant | 1 in 1050 | No cosmic structure |
| Gravity strength (force range) | 1 in 1034 | No life-sustaining planets |
| EM/gravity ratio | 1 in 1040 | No galaxies, stars, planets |
| Electron-to-proton ratio | 1 in 1037 | EM dominates gravity |
| Strong nuclear force | 2% tolerance | No elements heavier than hydrogen |
| Neutron-proton mass difference | ~1 part in 700 | No stellar fusion or no hydrogen |
| Fine-structure constant (alpha) | ~1% | Chemistry and fusion rewritten |
| Yukawa couplings (Barnes) | 1 in 1013 | No stable matter |
Penrose's Number: 1 in 1010123
What he calculated: The probability the universe would begin in the extraordinarily low-entropy state required for the Second Law of Thermodynamics and complex structures.
The number: 1 in 1010123
Writing this number requires 10123 zeroes. The observable universe contains only ~1080 particles. You could not write all the zeroes if each particle in the universe represented one zero.
The number: 1 in 1010123
Writing this number requires 10123 zeroes. The observable universe contains only ~1080 particles. You could not write all the zeroes if each particle in the universe represented one zero.
Combined Fine-Tuning Estimates
| Source | Combined Probability |
|---|---|
| Collins (conservative) | 1 in 1053 (cosmological constant alone) |
| Barnes (life-supporting planet) | 1 in 10390 |
| Barnes (functional proteome) | 1 in 10722,000 |
| Hugh Ross (922 parameters) | 1 in 101032 |
| Penrose (initial entropy) | 1 in 1010123 |
The Multiverse Doesn't Solve It
Boltzmann Brain Problem: An infinite multiverse produces infinitely more random "brain flickers" than ordered observers. We should be Boltzmann brains floating in chaos. We're not. Multiverse falsified by observation.
Measure Problem: Both normal observers and Boltzmann brains are infinite in a multiverse. No principled way to compare them. A theory that can't generate determinate probabilities can't explain anything.
Meta-Fine-Tuning: Any multiverse generator (eternal inflation, string landscape) requires its own fine-tuned parameters. The problem is displaced, not solved.
Measure Problem: Both normal observers and Boltzmann brains are infinite in a multiverse. No principled way to compare them. A theory that can't generate determinate probabilities can't explain anything.
Meta-Fine-Tuning: Any multiverse generator (eternal inflation, string landscape) requires its own fine-tuned parameters. The problem is displaced, not solved.
Resurrection: Swinburne's 97%
| Parameter | Value |
|---|---|
| P(God exists | natural theology) | 1/2 |
| P(God would become incarnate | God exists) | 1/2 |
| P(evidence | NOT God incarnate) | 1/1000 |
| P(resurrection | all evidence) | 0.97 (97%) |
McGrews' Bayes Factor: 1039 to 1
Timothy & Lydia McGrew (Blackwell Companion to Natural Theology, 2009): ~108 per independent witness testimony, multiplied across witnesses.
Combined Bayes factor: 1039 to 1 in favor of the resurrection.
That is: 1,000,000,000,000,000,000,000,000,000,000,000,000,000 to 1.
That is: 1,000,000,000,000,000,000,000,000,000,000,000,000,000 to 1.
Joseph Bae's calculation with 500 witnesses and barely-better-than-coinflip reliability: 99.999...% with 65 nines after the decimal.
Scholarly Consensus (Habermas, 3,400+ Publications)
| Fact | Scholar Agreement |
|---|---|
| Jesus existed as a historical person | ~99% |
| Jesus died by Roman crucifixion | ~99% |
| Disciples had experiences they believed were risen Jesus | ~95%+ |
| Paul converted after claiming risen Christ experience | ~95%+ |
| James (skeptical brother) converted | ~90%+ |
| The tomb was empty | ~75% |
| 1 Cor 15 creed: within 2-5 years of crucifixion | Near universal |
Godel's Proof: Computer Verified (2013)
Kurt Godel (greatest logician since Aristotle) wrote a formal modal-logic proof of God's existence.
In 2013, Christoph Benzmuller & Bruno Woltzenlogel Paleo formalized it in Isabelle/HOL and verified it with automated theorem provers (LEO-II, Satallax).
What they proved:
1. The axioms are consistent (no contradiction)
2. The derivation is logically valid (each step follows necessarily)
3. The conclusion — "necessarily, a God-like being exists" — follows from the axioms
No hidden logical errors exist in the proof. The computer confirmed it. The question is only whether you accept the axioms.
In 2013, Christoph Benzmuller & Bruno Woltzenlogel Paleo formalized it in Isabelle/HOL and verified it with automated theorem provers (LEO-II, Satallax).
What they proved:
1. The axioms are consistent (no contradiction)
2. The derivation is logically valid (each step follows necessarily)
3. The conclusion — "necessarily, a God-like being exists" — follows from the axioms
No hidden logical errors exist in the proof. The computer confirmed it. The question is only whether you accept the axioms.
Plantinga's Modal Argument (Valid in S5)
If it is even POSSIBLE that a maximally great being exists, then that being NECESSARILY exists.
Using S5 modal logic (where possible necessity = necessity):
1. It is possible that a maximally great being exists
2. If possible, then in some world W it is maximally excellent in ALL worlds
3. By S5: it is maximally excellent in the actual world
4. Therefore a maximally great being exists
Logically valid — uncontested by any logician. The debate is only about premise 1.
Using S5 modal logic (where possible necessity = necessity):
1. It is possible that a maximally great being exists
2. If possible, then in some world W it is maximally excellent in ALL worlds
3. By S5: it is maximally excellent in the actual world
4. Therefore a maximally great being exists
Logically valid — uncontested by any logician. The debate is only about premise 1.
Kalam + BGV Theorem
The syllogism:
- Everything that begins to exist has a cause
- The universe began to exist
- Therefore the universe has a cause
The BGV Theorem (Borde-Guth-Vilenkin, 2003): Any universe with average Hubble expansion > 0 must have a past boundary. Applies to inflationary models, cyclic models, and multiverse scenarios.
Vilenkin: "With the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe."
Vilenkin: "With the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe."
16 Modern Mathematical Provers of God
| Thinker | Field | Argument | Key Claim |
|---|---|---|---|
| Godel | Logic | Modal ontological | Computer-verified: necessarily God exists |
| Benzmuller | Computer Science | Automated verification | Godel's proof has no logical errors |
| Plantinga | Philosophy (Notre Dame) | Modal ontological (S5) | Possible maximal greatness → necessary existence |
| Langan | Autodidact (IQ 195+) | CTMU metaphysical | Reality as SCSPL entails G.O.D. |
| Swinburne | Philosophy (Oxford) | Bayesian cumulative | P(God|evidence) > 50% |
| Craig | Philosophy | Kalam cosmological | BGV theorem + causal principle |
| Spitzer | Physics / Jesuit | Physics-based | BGV + entropy + fine-tuning |
| Lennox | Mathematics (Oxford) | Teleological | Math intelligibility requires rational Creator |
| Dembski | Mathematics (U Chicago PhD) | Information-theoretic | Specified complexity → design. Bound: 10150 |
| Meyer | Phil. of Science (Cambridge PhD) | DNA information | 1 in 10164 for one functional protein |
| Feser | Philosophy | Aristotelian-Thomistic | 5 independent formal proofs (27-50+ steps) |
| Pruss | Math PhD + Phil. PhD (Baylor) | Leibnizian cosmological | PSR + BCCF → necessary being |
| Koons | Philosophy (UT Austin) | Mereological cosmological | Cosmos as aggregate requires necessary cause |
| Berlinski | Mathematics (Princeton PhD) | Critique of atheism | Materialism fails mathematically |
| Collins | Philosophy/Physics (Messiah U) | Fine-tuning Bayesian | P(fine-tuning|theism) >> P(fine-tuning|atheism) |
| Rutten | Philosophy (VU Amsterdam) | Modal-epistemic | If God's non-existence is unknowable, God necessarily exists |
Universal finding: The logic is valid in every case. No one has found a logical error in any of these proofs. The debates are always about the premises.
Manuscript Evidence: Unparalleled
| Text | Manuscripts | Time Gap |
|---|---|---|
| New Testament | 5,800 Greek / 25,000+ total | ~25-50 years |
| Homer's Iliad | ~1,757 | ~400 years |
| Plato | 7 | ~1,200 years |
| Caesar's Gallic Wars | 9-10 | ~950 years |
| Tacitus' Annals | 2 | ~1,000 years |
400,000 textual variants but <0.1% are both meaningful AND viable. Both Bart Ehrman (skeptic) and Daniel Wallace (evangelical) agree: zero doctrines affected by any variant.
Dead Sea Scrolls: 95%+ Accuracy Over 1,000 Years
The Great Isaiah Scroll (dated ~125 BC) compared to the Masoretic text (~930 AD): word-for-word identical in more than 95% of the text. The 5% = spelling variations and scribal slips. No meaning changes. Isaiah 53 (Suffering Servant): 166 words, only 17 letter differences — 10 spelling, 4 conjunctions, 3 = one word ("light"). Zero meaning changes.
Archaeological Confirmation
53 biblical persons confirmed by archaeological inscriptions (Mykytiuk, Purdue, published in Biblical Archaeology Review).
"No archaeological discovery has ever controverted a biblical reference. Scores of findings confirm in clear outline or exact detail historical statements in the Bible." — Nelson Glueck, Reform Jewish scholar, discoverer of 1,500+ ancient sites
Key confirmations: Tel Dan Inscription (House of David), Pilate Stone, Cyrus Cylinder, Mesha Stele.
Faith & Health: The Empirical Data
| Outcome | Effect of Religious Practice | Source |
|---|---|---|
| Suicide | 5x lower rate | Harvard, N=89,708 |
| Deaths of despair | 68% lower (women), 33% (men) | Harvard |
| All-cause mortality | 33% reduction | JAMA Internal Medicine, N=74,534 |
| Life expectancy | +7 to +14 years | Multiple studies |
| Depression | 67% of rigorous studies show inverse | Koenig, 500+ papers |
| Addiction (AA vs CBT) | 42% vs 35% abstinence | Cochrane Review (gold standard) |
| Divorce | 31-35% lower | Wilcox, UVA |
| Happiness | 44% more "very happy" | Pew Research |
| Prosocial behavior | r=.13 across 811,663 participants | Meta-analysis |
If a pharmaceutical drug produced these effects across this many domains simultaneously, it would be the most prescribed medication in human history.
Scientists Who Believe
36% of elite scientists believe in God (Ecklund, Rice University, 1,700 professors at 21 elite universities). Christians have won 72.5% of Chemistry Nobels, 65.3% of Physics Nobels, 78.3% of Peace Nobels. Newton, Faraday, Maxwell, Pasteur, Mendel, Kepler, Pascal, Boyle, Lord Kelvin — all Bible-believing Christians.
Christianity vs. Other Religions: The Data
| Category | Christianity | Nearest Competitor |
|---|---|---|
| Manuscripts | 25,000+ | Homer: 1,757 |
| Time to earliest copy | ~25-50 years | Buddha: ~500+ years |
| Resurrection claim with named eyewitnesses | 500+ cited by Paul within 25 years | No other religion claims this |
| Specific fulfilled prophecy | 191-456 messianic prophecies | No comparable system exists |
| Growth under persecution (no political power) | 40%/decade for 300 years | Islam grew through military conquest |
| Hostile witness conversions in founding | Paul (persecutor), James (skeptic) | No parallel in other religions |
| Creed dating | 2-5 years after events | Buddha's texts: 200-500 years later |
Growth Under Fire (Rodney Stark, Princeton)
~1,000 Christians in 40 AD → ~33 million by 350 AD
40% per decade for 300 years. Under active persecution. Without political power. Without military force. Without wealth. Against a hostile cultural establishment.
No other religion in history has achieved this combination.
40% per decade for 300 years. Under active persecution. Without political power. Without military force. Without wealth. Against a hostile cultural establishment.
No other religion in history has achieved this combination.
Hostile Witness Conversions
Paul: Pharisee, active persecutor, oversaw Stephen's execution → chief missionary, endured beatings/imprisonments/execution. His genuine letters are the earliest Christian documents.
James: Jesus' brother, skeptic during ministry (Mark 3:21, John 7:5) → leader of Jerusalem church → martyred 62 AD (attested by Josephus, non-Christian).
People die for beliefs they hold sincerely. People do not knowingly die for events they invented. If the resurrection was a hoax, the apostles knew it — and at least four (James son of Zebedee, Peter, Paul, James brother of Jesus) went to their deaths rather than recant.
The Bayesian Cumulative Case
Swinburne (Oxford): ~11 evidential factors ratcheted through Bayes' theorem. Each C-inductive (raises probability). Together P-inductive (crosses 50%). Religious experience is the tipping factor.
Unwin (physicist): Starting from 50% prior, 6 evidence factors → 67%.
The math of cumulation: If 5 independent arguments each raise probability by 1.5x, combined = 1.55 = 7.59x shift. Even modest individual arguments produce massive cumulative force.
Honest Criticism
The numbers depend on priors. Change Swinburne's P(God)=1/2 to P(God)=10-6 and the cumulative case collapses. There is no objective way to determine the prior probability of God's existence. The Bayesian machinery is formally valid but the inputs are subjective.
However: the SAME objection applies to rejecting God. Setting a prior of 10-6 is itself a subjective judgment, not an empirical measurement. Everyone starts with a prior. The question is which prior is more reasonable.
However: the SAME objection applies to rejecting God. Setting a prior of 10-6 is itself a subjective judgment, not an empirical measurement. Everyone starts with a prior. The question is which prior is more reasonable.
The Verdict
The evidence does not merely invite faith. It demands an explanation.
As Langan says: "The existence of God isn't just faith — it's logic."
As Swinburne concludes: "On the total evidence, the probability that God exists is greater than 0.5."
As the health data shows: If faith were a drug, it would be the most prescribed medication in history.
The numbers are in. What you do with them is between you and God.
- Prophecy: 1 in 1017 for 8 prophecies. 1 in 10157 for 48. Daniel predicted Palm Sunday to the day.
- Fine-tuning: 1 in 1010123 for initial entropy. The multiverse doesn't solve it.
- Resurrection: 97% (Swinburne). 1039:1 Bayes factor (McGrews). ~99% of scholars: Jesus existed and was crucified. ~95%: disciples believed they saw Him risen.
- Formal proofs: Godel's proof computer-verified. Plantinga's valid in S5. 16 mathematicians/logicians have produced formal arguments. Zero logical errors found in any.
- Manuscripts: 25,000+ copies. 95%+ accuracy over 1,000 years. 53 persons confirmed archaeologically. Zero doctrines affected by variants.
- Health: 5x lower suicide. 33% lower mortality. +7-14 years of life. Outperforms CBT for addiction.
- Uniqueness: No other religion has this combination of early manuscript evidence, fulfilled prophecy, hostile witness conversions, growth under persecution, and civilizational impact.
As Langan says: "The existence of God isn't just faith — it's logic."
As Swinburne concludes: "On the total evidence, the probability that God exists is greater than 0.5."
As the health data shows: If faith were a drug, it would be the most prescribed medication in history.
The numbers are in. What you do with them is between you and God.